sign in sign up
<<
Home , Cell cycle checkpoint, Cyclin A, Cyclin B, MYT1, PIN1, Plk1

(2013) 32, 4778–4788 & 2013 Macmillan Publishers Limited All rights reserved 0950-9232/13 www.nature.com/onc

ORIGINAL ARTICLE The CDK1 inhibitory MYT1 in DNA damage checkpoint recovery

JPH Chow and RYC Poon

Inhibition of -dependent kinase 1 (CDK1) by is a key regulatory mechanism for both the unperturbed and the DNA damage checkpoint. Although both and MYT1 can phosphorylate CDK1, little is known about the contribution of MYT1. We found that in contrast to WEE1, MYT1 was not important for the normal cell cycle or checkpoint activation. Time-lapse microscopy indicated that MYT1 did, however, have a rate-determining role during checkpoint recovery. Depletion of MYT1 induced precocious mitotic entry when the checkpoint was abrogated with inhibitors of either CHK1 or WEE1, indicating that MYT1 contributes to checkpoint recovery independently of WEE1. The acceleration of checkpoint recovery in MYT1-depleted cells was due to a lowering of threshold for CDK1 activation. The kinase activity of MYT1 was high during checkpoint activation and reduced during checkpoint recovery. Importantly, although depletion of MYT1 alone did not affect long- term , it potentiated with DNA damage to inhibit cell growth in clonogenic survival and tumor xenograft models. These results reveal the functions of MYT1 in checkpoint recovery and highlight the potential of MYT1 as a target for anti- therapies.

Oncogene (2013) 32, 4778–4788; doi:10.1038/onc.2012.504; published online 12 November 2012 Keywords: cell cycle; checkpoint; ionizing radiation

INTRODUCTION CHK1/CHK2, which in turn inactivates (reviewed in Chen 15 Cyclin-dependent kinase 1 (CDK1) is one of the major for and Poon ). This response tips the balance of the CDK1 Thr14/ driving cells into . Its activation requires the binding of Tyr15 phosphorylation toward inactivation of the kinase. Indeed, and the phosphorylation of the T-loop. Cyclin B starts to the checkpoint induced by ionizing radiation (IR) can be abrogated 16 accumulate from and is destroyed at the end of mitosis.1 by overexpressing a non-phosphorylatable CDK1 mutant. Before mitosis, cyclin B–CDK1 complexes are inactivated through Once damaged DNA is repaired, the checkpoint has to be phosphorylation of Thr14 and Tyr15 by WEE1 and MYT1.2,3 While switched off to restart the cell cycle. This requires CDC25B, but not WEE1 specifically phosphorylates Tyr15, MYT1 shows a preference CDC25A and CDC25C, indicating a divergence of the roles of cell for Thr14. cycle components in checkpoint recovery and unperturbed cell 4 Just before mitosis, WEE1 and MYT1 are downregulated to allow cycle. Similarly, although is not required for normal mitotic 4 CDK1 activation. The NH2-terminal region of WEE1 is phosphory- entry, it is important for checkpoint recovery. While WEE1 is lated by CDK1, PLK1 and CK2, creating a phosphodegron for known to have an important role in the G2 DNA damage 17 SCFbTrCP-dependent degradation.4–6 Any residual WEE1 is also checkpoint, next to nothing is known about the function of inhibited by CDK1- and AKT/PKB-dependent phosphorylation as MYT1 in the checkpoint. well as by binding to PIN1.7 By contrast, relatively little is known In this study, we have investigated the functions of MYT1 in about how MYT1 is inactivated. In mammalian cells, the checkpoint recovery. We found that while MYT1 was not essential inactivation of MYT1 coincides with its phosphorylation by PLK1 for the unperturbed cell cycle or the activation of the DNA damage and CDK1.2,8,9 checkpoint, it had a rate-determining function in checkpoint WEE1 has an indispensable role in controlling the timing of recovery. Downregulation of MYT1 also promoted DNA damage- mitosis during normal cell cycle. Depletion of WEE1 resulted in mediated cytotoxicity in cell line and nude mice models. premature condensation.10 By contrast, MYT1 appears to have a relatively minor role in the somatic cell cycle.11 As MYT1 is the only CDK1 inhibitory kinase in - RESULTS arrested oocytes, MYT1 is believed to have an important MYT1 is not involved in the control of the unperturbed cell cycle 12 role in the meiotic cycle during early development. Under this and activation of the G2 DNA damage checkpoint condition, MYT1 is inactivated by XRINGO- and CDK-mediated We started by analyzing the temporal expression pattern of MYT1 phosphorylation.13 In starfish, MYT1 activity is downregulated by in HeLa cells. Cells were released from a double thymidine AKT-dependent phosphorylation during .14 synchronization procedure and harvested every 3 h. Mitosis Proper cell cycle progression requires several checkpoints that occurred at between 9–12 h, as indicated by the DNA contents Tyr15 regulate the activities of CDKs. The G2 DNA damage checkpoint and the expression of phosphos-CDK and mitotic exerts its effect mainly through the inhibitory phosphorylation of (Figure 1a). While the expression of WEE1 decreased during CDK1. Upon DNA damage, ATM/ATR phosphorylates and activates mitosis, MYT1 displayed a dramatic gel mobility shift. Addition of

Division of Life Science and Center for Cancer Research, Hong Kong University of Science and Technology, Kowloon, Hong Kong. Correspondence: Professor RYC Poon, Division of Life Science, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong. E-mail: [email protected] Received 10 May 2012; revised 14 September 2012; accepted 18 September 2012; published online 12 November 2012 MYT1 in checkpoint recovery JPH Chow et al 4779

Figure 1. MYT1 is not important for the unperturbed cell cycle. (a) Level and phosphorylation of MYT1 during the cell cycle. HeLa cells were synchronized with a double thymidine block method. At different time points after released from the block, the cells were harvested and analyzed with flow cytometry (the positions of 2N and 4N DNA contents are indicated). Lysates were also prepared and the expression of the indicated was analyzed with immunoblotting. Uniform loading of lysates was confirmed by immunoblotting for actin. The positions of the phosphorylated and unmodified MYT1 are indicated. (b) Gel mobility shift of MYT1 during mitosis is phosphorylation-dependent. MYT1 was immunoprecipitated from mitotic cell extracts (from cells collected at 12 h after release from double thymidine block as described in panel a). The immunoprecipitates were treated with either buffer or lPPase. The gel mobility of MYT1 was then analyzed with immunoblotting. (c) MYT1 is not required for the activation of various checkpoints. HeLa cells were either mock-transfected or transfected with siRNA against MYT1 (siMYT1). After 48 h, the cells were treated with hydroxyurea (HU), IR or nocodazole (NOC) for 16 h. Lysates were then prepared and the expression of the indicated proteins was analyzed with immunoblotting. Equal loading of lysates was confirmed by immunoblotting for actin. (d) MYT1 is not required for checkpoint-induced cell cycle arrest. Cells were transfected and treated as in panel c. The cells were then fixed and analyzed with flow cytometry. (e) MYT1 depletion does not affect the timing of mitotic entry. HeLa cells expressing histone H2B-GFP were transfected with siRNAs against MYT1, PLK1 or WEE1. The cells were synchronized with a double thymidine procedure and released into the cell cycle. Time-lapse microscopy was used to analyze the time taken by individual cells to enter mitosis (mean±90% confidence coefficient; n ¼ 30). The original data of the individual cells are shown in Supplementary Figure S2. the inhibitor MG132 stabilized WEE1 and slower-migrating form of MYT1 after lambda without affecting MYT1 (Supplementary Figure S1A), verifying that treatment verified that MYT1 was strongly phosphorylated during MYT1 was not regulated by degradation. The disappearance of the mitosis (Figure 1b).9

& 2013 Macmillan Publishers Limited Oncogene (2013) 4778 – 4788 MYT1 in checkpoint recovery JPH Chow et al 4780 To study the functions of MYT1, we depleted MYT1 with small indicate that unlike that for WEE1, there is no evidence that MYT1 interfering RNAs (siRNAs) (siMYT1) and verified the depletion with has a role in the timing of mitosis during the unperturbed cell immunoblotting (Figure 1c). This resulted in negligible effects on cycle of HeLa cells. normal cell cycle distribution (Figure 1d). MYT1-depleted cells were able to be trapped by hydroxyurea, IR or nocodazole in S phase, or mitosis, respectively, indicating that the proper MYT1 is a component of the checkpoint recovery network activation of the various checkpoints did not require MYT1. Time- Several potential cell cycle regulators including PLK1 and CDC25B lapse microscopy also confirmed DNA damage was able to induce appear to be non-essential for normal G2–M transition, yet have a cell cycle arrest in MYT1-depleted cells (see below). important roles during the recovery of the G2 DNA damage WEE1 prevents premature activation of cyclin–CDK1 complexes checkpoint.19 To determine whether MYT1 is involved in during G2 phase. To evaluate the possible contribution of MYT1 on checkpoint recovery, siMYT1 was transfected before DNA the timing of mitosis, cells expressing histone H2B-GFP were damage was introduced with IR. HeLa cells were used as a generated and transfected with siMYT1. After they were model because IR exclusively activated the G2 DNA damage synchronously released from a double thymidine block, individual checkpoint in these cells ( is degraded by HPV E6). We used a cells were then tracked with time-lapse microscopy to monitor the CHK1 inhibitor (UCN-01) to turn off the checkpoint, thereby timing of mitosis (Supplementary Figure S2). The average time mimicking checkpoint recovery and inducing mitosis taken for the control cells to enter mitosis was B600 min synchronously in the population.20 Time-lapse microscopy (Figure 1e). This was dramatically reduced to B300 min confirmed that almost no cell underwent mitosis after IR when WEE1 was downregulated with siRNA. In contrast, MYT1- treatment (Supplementary Figure S3A). Moreover, MYT1 depletion depleted cells entered mitosis at a similar time as control cells. did not affect the IR-mediated cell cycle arrest over the imaging PLK1 is implicated to have a role in mitotic entry.18 Interestingly, period (Supplementary Figure S3B). Addition of UCN-01 stimu- depletion of PLK1 (either alone or together with MYT1) did not lated mitosis (Figure 2a, Supplementary Figure S4). While mitotic affect the timing of mitotic entry (Figure 1e). Instead, knockdown entry occurred at B250 min in control cells, it was shortened by of PLK1 abolished , resulting in a very protracted 50 min in MYT1-depleted cells. As a positive control, depletion of mitosis (Supplementary Figure S2). Taken together, these results PLK1 significantly delayed mitotic entry during checkpoint

Figure 2. MYT1 is a component of the checkpoint recovery network. (a) Depletion of MYT1 accelerates entry into mitosis during checkpoint recovery. HeLa cells stably expressing histone H2B-GFP were transfected with siRNAs against control, MYT1, and/or PLK1. After incubation for 24 h, the cells were irradiated with 15 Gy of IR. After 16 h, the cells were treated with either buffer or UCN-01, before subjected to time-lapse microscopy for 22 h. The average mitotic entry time was measured (mean±90% confidence coefficient). Statistical analysis was performed with Student’s t-test. The raw data of individual cells are shown in panel b and Supplementary Figure S4. (b) MYT1 depletion promotes mitotic catastrophe after checkpoint recovery. Cells were transfected with either control or siMYT1, subjected to IR and UCN-01 treatment, before analyzed with time-lapse microscopy exactly as described in panel a. Each horizontal line represents one cell. gray, ; black, mitosis (from DNA condensation to or cell death); truncated bars, cell death. The time of cell death after mitosis is defined by the death of one of the daughters.

Oncogene (2013) 4778 – 4788 & 2013 Macmillan Publishers Limited MYT1 in checkpoint recovery JPH Chow et al 4781 recovery4 (Figure 2a). Interestingly, the delay caused by PLK1 depletion was reversed by co-depletion of MYT1, suggesting a downstream role of MYT1. Checkpoint-abrogated cells frequently underwent cell death during mitosis. Alternatively, some cells were able to complete the mitosis (albeit frequently with failure), then either survived or died (postmitotic cell death) in the following interphase.20 We found that siMYT1 increased mitotic cell death following checkpoint abrogation (Figure 2b). By the end of the imaging period (22 h after UCN-01 treatment), only 20% of MYT1- depleted cells remained (compare with over 40% in the control). As siMYT1 alone was not cytotoxic (Supplementary Figure S2C), the increase in mitotic cell death was likely to be caused by the shortening of mitotic entry time after checkpoint abrogation. The premature checkpoint recovery in MYT1-depleted cells was further verified using flow cytometry (Figure 3a). As expected, IR treatment arrested cells with DNA contents corresponding to G2/M phase. Bypass of the checkpoint with UCN-01 promoted the appearance of G1 cells. This was further accelerated in MYT1-depleted cells and delayed in PLK1-depleted cells. The sub-G1 population in PLK1-depleted cells was consistent with the extensive mitotic cell death observed using time-lapse microscopy (Supplementary Figure S4). To examine the accumulation of the mitotic markers during checkpoint recovery, nocodazole was added together with UCN- 01 to prevent checkpoint-abrogated cells from exiting mitosis. The fact that depletion of MYT1 promoted checkpoint recovery was supported by the premature accumulation of phospho-histone H3Ser10 and disappearance of phospho-CDK1Tyr15 (Figure 3b). Notably, the absence of MYT1 resulted in a lower level of CDK1Tyr15 phosphorylation during checkpoint activation (see later). As a control, PLK1 depletion delayed CDK1Tyr15 depho- sphorylation and histone H3Ser10 phosphorylation. We next repeated the checkpoint recovery experiments with a siRNA that targeted a different region of MYT1 (Supplementary Figure S5). Entry into mitosis after checkpoint abrogation was also speeded up by the second siMYT1, supporting the specificity of the effects of MYT1 knockdown. To exclude the possibility that the function of MYT1 is only limited to HeLa cells, we also demonstrated similarly accelerated mitotic entry following check- point abrogation in MYT1-depleted U2OS cells (Supplementary Figure S6). Interestingly, while depletion of PLK1 delayed mitosis, it was not reversed by the co-depletion of MYT1, suggesting that PLK1 probably have multiple targets during checkpoint recovery in U2OS cells. Finally, we performed the converse experiment by overexpressing MYT1. As expected, the mitotic entry time after checkpoint abrogation was significantly delayed after MYT1 overexpression (Supplementary Figure S7). Figure 3. Depletion of MYT1 accelerates checkpoint recovery. (a) MYT1 depletion promotes the appearance of G1 cells after Taken together, our results indicate that although MYT1 is not checkpoint abrogation. HeLa cells were transfected with siRNAs essential for normal cell cycle and arrest following DNA damage, it against control, MYT1, and/or PLK1. After incubation for 24 h, the has a rate-determining role for the recovery of the G2 DNA cells were irradiated with IR. After 16 h, the cells were treated with damage checkpoint. UCN-01. The cells were harvested at the indicated time points (after UCN-01 addition) and analyzed with flow cytometry. (b)MYT1 depletion promotes CDK1 activation and mitosis after checkpoint MYT1 and WEE1 contributes independently to checkpoint abrogation. HeLa cells were transfected and treated with IR and recovery UCN-01 as in panel a. Nocodazole was added at the same time as As UCN-01 inhibited upstream activators for both MYT1 and WEE1, UCN-01 to trap checkpoint-abrogated cells in mitosis. At the indicated time points (after UCN-01 addition), the cells were we next investigated the contribution of MYT1 and WEE1 more harvested and analyzed with immunoblotting. directly with a specific WEE1 inhibitor MK1775.21 Inhibition of WEE1 was sufficient to induce checkpoint abrogation (Figure 4a). Inhibition of WEE1 promoted a faster mitotic entry (B160 min, Figure 4b) than the inhibition of CHK1 (B250 min, Figure 2a), a H3Ser10 after MK1775 treatment (Figure 4c). Collectively, these data more upstream checkpoint component. Interestingly, more indicate that WEE1 and MYT1 contribute independently to the checkpoint-abrogated cells induced by MK1775 underwent timing of checkpoint recovery. mitotic catastrophe than those induced by UCN-01. More significantly, depletion of MYT1 further accelerated mitotic entry MYT1 is involved in setting a threshold for CDK1 activation during following MK1775 treatment (Figures 4a and b). Immunoblotting checkpoint recovery analysis further confirmed that MYT1 depletion promoted faster To understand how MYT1 affects checkpoint recovery, we first of CDK1Try15 and phosphorylation of histone excluded the possibility that the degree of DNA damage was

& 2013 Macmillan Publishers Limited Oncogene (2013) 4778 – 4788 MYT1 in checkpoint recovery JPH Chow et al 4782

Figure 4. Both MYT1 and WEE1 contribute to the control of checkpoint recovery. (a) MYT1 depletion promotes MK1775-mediated checkpoint recovery. HeLa cells stably expressing histone H2B-GFP were transfected with either control or siMYT1. After incubation for 24 h, the cells were irradiated with 15 Gy of IR. After 16 h, the cells were treated with MK1775 and analyzed with time-lapse microscopy (n ¼ 50). Each horizontal line represents one cell. gray, interphase; black, mitosis (from DNA condensation to anaphase or cell death); truncated bars, cell death. The time of cell death after mitosis is defined by the death of one of the daughters. (b) The average mitotic entry time in panel a was measured (mean±90% confidence coefficient). (c) Depletion of MYT1 accelerates CDK1 dephosphorylation and mitosis after checkpoint abrogation. HeLa cells were transfected with either control or siMYT1. After incubation for 24 h, the cells were irradiated with 15 Gy of IR. After 16 h, the cells were treated with MK1775 and analyzed with immunoblotting. Equal loading of lysates was confirmed by immunoblotting for actin.

actually affected by MYT1. Similar number of g-H2AX foci was depletion (Figure 5c). Although histone H3Ser10 phosphorylation found after IR treatment in control, MYT1- or PLK1-depleted cells did not increase in MYT1-depleted cells (indicating that the (Figure 5a), suggesting that the extent of DNA damage was not checkpoint was not bypassed), there was a two-fold increase in affected by MYT1 depletion. CDK1 kinase activity over control-irradiated cells. This was not due Given that CDK1Tyr15 phosphorylation appears to be compro- to a change in the total level of (Figure 5b). This was also mised in MYT1-depleted cells (Figure 3b), we next examined the supported by the similar maximum CDK1 activity attained when activity of CDK1. As a control, knockdown of WEE1 uncoupled the cyclin B1–CDK1 complexes were completely activated with DNA damage checkpoint, resulting in CDK1Tyr15 dephosphoryla- recombinant CDC25A (Figure 5c). During checkpoint recovery, tion and histone H3Ser10 phosphorylation (Figure 5b). Indeed, the although the CDK1 activity reached similar maximum levels in the kinase activity of CDK1 was substantially increased after WEE1 presence or absence of MYT1, it reached a higher level faster after

Oncogene (2013) 4778 – 4788 & 2013 Macmillan Publishers Limited MYT1 in checkpoint recovery JPH Chow et al 4783

Figure 5. MYT1 is involved in checkpoint-mediated CDK1 inactivation. (a) The extent of DNA damage is not affected by MYT1 depletion. HeLa cells were mock-transfected or transfected with siRNA against MYT1 or PLK1. After 24 h, the cells were treated with IR and grown for another 16 h. The cells were then fixed and subjected to immunostaining against g-H2AX and counterstained with Hoechst 33258. Representative confocal microscopy images are shown. The average number of g-H2AX foci was quantified (n ¼ 30). (b) The G2 DNA damage checkpoint cannot be maintained in the absence of WEE1. HeLa cells were transfected with control, siMYT1, siPLK1 or siWEE1. After 24 h, the cells were treated with IR and incubated for another 16 h before harvested. Lysates were prepared and analyzed with immunoblotting. (c) Depletion of MYT1 leads to some increase in CDK1 activity after DNA damage. Cells were transfected with siRNAs and irradiated as in panel b. Lysates were subjected to immunoprecipitation with antibodies against CDK1. The immunoprecipitates were then incubated with either buffer or purified GST-CDC25A as described in Materials and Methods. The kinase activity was then assayed using histone H1 as a substrate. Phosphorylation was quantified with a PhosphorImager (mean±s.d. of three independent experiments). (d) Full activation of CDK1 after checkpoint abrogation is achieved earlier in MYT1-depleted cells. HeLa cells were transfected with either control or siMYT1. After 24 h, the cells were treated with IR and incubated for another 16 h. The cells were then treated with UCN-01 and harvested at the indicated time points. Lysates were subjected to immunoprecipitation with antibodies against CDK1. The kinase activity was then assayed using histone H1 as a substrate. Phosphorylation was quantified with a PhosphorImager (mean±s.d. of three independent experiments).

MYT1 depletion (Figure 5d). This may provide a molecular siMYT1 alone did not affect clonogenic survival. When combined explanation on the faster checkpoint recovery in the absent of with IR treatment, however, siMYT1 exerted a pronounced effect MYT1. on survival. Hence although MYT1 is not required for the Collectively, these results indicate that the acceleration of immediate cell cycle arrest after DNA damage (up to 36 h, see checkpoint recovery in MYT1-depleted cells is not due to a above), it is important for maintaining long-term cell survival after difference in the extent of DNA damage, but is likely to be caused DNA damage. by a decrease in CDK1 inhibitory phosphorylation during We further tested the growth of MYT1-depleted cells with nude checkpoint activation, with a consequent faster activation of mouse xenograft models. Initially, we treated siMYT1-transfected CDK1 during recovery. HeLa cells with IR in vitro before injecting them subcutaneously into nude mice. When the cells were not treated with IR, no significant difference in the increase in tumor volume was found Depletion of MYT1 potentiates with DNA damage to inhibit tumor between control and siMYT1-transfected cells (Figure 6b). When IR growth in mouse xenograft models was delivered, however, tumor growth was significantly delayed in Given that MYT1-depeleted cells underwent premature check- siMYT1-transfected cells (Figure 6c). point recovery after DNA damage, we next examined their We next performed experiments in which the cells were first long-term growth potential (Figure 6a). Unlike siWEE1 or siPLK1, injected into the nude mice before IR was delivered in vivo.

& 2013 Macmillan Publishers Limited Oncogene (2013) 4778 – 4788 MYT1 in checkpoint recovery JPH Chow et al 4784

Figure 6. MYT1 is important for long-term survival and growth after DNA damage in mouse xenograft models. (a) Depletion of MYT1 reduces clonogenic survival after DNA damage. HeLa cells were transfected with siRNAs against control, PLK1, MYT1 or WEE1. After 24 h, the cells were either untreated (upper panel) or treated with 2 Gy of IR (lower panel). The cells were then seeded for colony formation assays (average of two independent experiments). (b) Depletion of MYT1 alone does not affect tumor growth in nude mice. HeLa cells were either mock-transfected or transfected with siMYT1. After 16 h, the cells were injected subcutaneously into nude mice. The volume of the tumor was measured on different days (mean±s.d., n ¼ 6). (c) Depletion of MYT1 with DNA damage in vitro reduces tumor growth in nude mice. HeLa cells were either mock-transfected or transfected with siMYT1. After 16 h, the cells were treated with 2 Gy of IR and injected subcutaneously into nude mice. The volume of the tumor was measured on different days (mean±s.d., n ¼ 6). (d) Depletion of MYT1 potentiates with irradiation to inhibit tumor growth in mouse xenograft. HeLa cells were either mock-transfected or transfected with siMYT1. After 16 h, the cells were injected subcutaneously into nude mice. After 24 h, the mice were irradiated with 2 Gy of IR. The volume of the tumor was measured on different days (mean±s.d., n ¼ 3). The mock (white arrows) and siMYT1-transfected cells (black arrows) were injected into two different sides of the same mice.

After irradiation, the growth of siMYT1-transfected cells was process. We developed an MYT1 kinase assay using cyclin B1– reduced pronouncedly (Figure 6d). The results were confirmed CDK1 complexes as substrates. As MYT1 is a membrane-associated using a second siRNA to downregulate MYT1 (Supplementary kinase, its activity cannot be easily assayed using standard Figure S8). Collectively, these results indicate that although MYT1 immunoprecipitation. The fact that WEE1 and MYT1 are located in depletion alone does not affect long-term cell growth, it different cellular compartments enabled us to separate them potentiates with DNA damage to inhibit growth in using fractionation (Figure 7a). MYT1 was present in the and xenograft models. membrane fraction but not in the cytosol and nuclear fraction. Conversely, WEE1 was detected only in the cytosol and nuclear fraction. Cyclin B1–CDK1 complexes were immunoprecipitated MYT1 is inactivated during checkpoint recovery from mitotic cell extracts and eluted with FLAG peptides (the Given the importance of MYT1 in checkpoint recovery, we next cyclin B1 is FLAG-tagged) to serve as a substrate for WEE1 and investigated whether the activity of MYT1 is regulated during the MYT1. As WEE1 and MYT1 are believed to be regulated by CDK1,

Oncogene (2013) 4778 – 4788 & 2013 Macmillan Publishers Limited MYT1 in checkpoint recovery JPH Chow et al 4785

Figure 7. MYT1 is inactivated during checkpoint recovery. (a) Fractionation of WEE1 and MYT1. At 16 h after irradiation, HeLa cells were treated with UCN-01 and harvested at the indicated time points. The cells were fractionated into membrane (mem) and cytosol and nucleus (cyt & nuc) fractions and analyzed with immunoblotting. (b) MYT1 is activated during checkpoint activation. At 16 h after irradiation, HeLa cells were treated with UCN-01 and harvested after 9 h. Mitotic cells were also collected after treatment with nocodazole for 16 h. Membrane fractions containing MYT1 were isolated and incubated with cyclin-B1–CDK1 complexes prepared as described in Materials and Methods. Phosphorylation of CDK1 was detected with a PhosphorImager. The positions of molecular size markers (in kDa) are indicated. (c) Checkpoint- activated MYT1 can be inactivated by PLK1. HeLa cells were treated with IR and harvested after 16 h. Membrane fractions containing MYT1 were isolated and incubated with GST or GST-PLK1. The kinase activity against cyclin-B1–CDK1 complexes was assayed. Phosphorylation of CDK1 was detected with a PhosphorImager. (d) MYT1 is inactivated during checkpoint recovery. HeLa cells were transfected with either control or siMYT1. After 24 h, the cells were treated with IR and incubated for another 16 h. The cells were then treated with UCN-01 and harvested at the indicated time points. Membrane fractions containing MYT1 were isolated and incubated with cyclin-B1–CDK1 complexes. Phosphorylation of CDK1Thr14 was detected with immunoblotting. (e) MYT1 is inactivated during checkpoint recovery. At 16 h after irradiation, HeLa cells were treated with UCN-01 and harvested at the indicated time points. Membrane fractions containing MYT1 were isolated and incubated with cyclin-B1–CDK1 complexes. Phosphorylation of CDK1 was detected with a PhosphorImager. (f) Similar expression of MYT1 in checkpoint-activated and checkpoint-abrogated cells. At 16 h after irradiation, HeLa cells were treated with UCN-01 and harvested after 9 h. Mitotic cells were also collected after treatment with nocodazole for 16 h. Membrane fractions containing MYT1 were isolated as described in Materials and Methods and incubated with or without lPPase. MYT1 was detected with immunoblotting. As multiple phosphorylated forms of MYT1 were present, the lPPase treatment allowed a more accurate comparison of MYT1 expression in different samples. we first incubated the cyclin B1–CDK1 complexes with a specific The relatively high MYT1 kinase activity after DNA damage was CDK1 inhibitor RO3306 (Supplemenatry Figure S9A). To assay the reduced after incubation with UCN-01, indicating that MYT1 was activity of MYT1, the RO3306-inhibited cyclin B1–CDK1 complexes inactivated during checkpoint recovery (Figures 7b and e). This were incubated with the MYT1 fraction. Phosphorylation of CDK1 was not simply due to a reduction in the total level of MYT1 by MYT1 was detected by using either radiolabelled ATP or (Figure 7f). Finally, we found that the activity of exogenously immunoblotting. As expected, MYT1 kinase activity was active expressed FLAG-tagged MYT1 was similarly reduced during during IR-mediated G2 DNA damage checkpoint and reduced checkpoint recovery (Supplementary Figure S9C). Collectively, during mitosis (Figure 7b). Furthermore, the MYT1 activity could these data indicate that MYT1 was turned off during checkpoint be abolished by incubation with recombinant PLK1 (Figure 7c), recovery, supporting the hypothesis that MYT1 has an important supporting the idea that PLK1 is a negative regulator of MYT1.9 role in the process. The MYT1-dependent phosphorylation of CDK1 was confirmed by using phospho-CDK1Thr14 antibodies (Figure 7d). Importantly, the CDK1Thr14 phosphorylation was significantly reduced when DISCUSSION MYT1 was first downregulated with siMYT1, indicating the Inhibitory phosphorylation of CDK1 is now firmly established as a specificity of the kinase assay for MYT1. As a control, mixing of key mechanism for normal cell cycle control as well as during DNA the cyclin B1-CDK1 substrate with the membrane fractions damage and replication checkpoints. Recently, we found that without incubation resulted in a very low background of CDK1Thr14 inhibitory phosphorylation of CDK1 may also have a role in mitotic phosphorylation, indicating that both the MYT1 fractions and exit, in particular when the degradation of the mitotic cyclins is CDK1 substrates contained only low basal CDK1Thr14 phosphoryla- compromised.22 The available evidence suggests that WEE1 has tion (Supplementary Figure S9B). an indispensable role in regulating CDK1. For example, while the

& 2013 Macmillan Publishers Limited Oncogene (2013) 4778 – 4788 MYT1 in checkpoint recovery JPH Chow et al 4786 expression of the non-phosphorylatable mutant of CDK1Tyr15 induces premature mitotic events, a similar mutant of CDK1Thr14 (a site phosphorylated by MYT1, but not WEE1) does not induce premature mitosis.23 In agreement with a major role of WEE1 in CDK1 regulation, expression of siWEE1 induced premature mitosis (Figure 1e) followed by massive cell death (Supplementary Figure S2). The DNA damage checkpoint was also abrogated by either siWEE1 or the WEE1 inhibitor MK1775, leading to precocious activation of CDK1 (Figure 5c), CDK1Tyr15 dephosphorylation and histone H3Ser10 phosphorylation (Figure 5b), mitotic entry (Figure 4a) and cell death (Figures 4a and 6a). By contrast, the contribution of MYT1 to cell cycle control appears to be relatively minor. Depletion of MYT1 affected neither the cell cycle distribution (Figure 1d) nor the timing of mitosis (Figure 1e, Supplementary Figure S2). The absence of MYT1 also did not have a strong impact on the activation of the DNA damage checkpoint. MYT1-depleted cells were able to arrest in G phase after irradiation, as indicated by 2 Figure 8. MYT1 is involved in checkpoint recovery. The G DNA flow cytometry (Figure 1d) or time-lapse microscopy 2 damage checkpoint maintains CDK1 in a Thr14- and Tyr15- (Supplementary Figure S3). Our results differ from a report that phosphorylated, inactive state. This is achieved through the indicates that Adriamycin-induced G2 arrest in HeLa cells is combined action of the activation of WEE1 and inhibition of the 24 abrogated when MYT1 is downregulated. Nevertheless, we CDC25 family. In this study, we show that although MYT1 is not found that downregulation of MYT1 did have a significant impact essential for checkpoint activation, it does contribute to the levels on the effectiveness of the G2 DNA damage checkpoint. IR- CDK1 inhibitory phosphorylation attained during checkpoint induced Thr14/Tyr15 phosphorylation was reduced after MYT1 activation. During checkpoint recovery, the signals from the ATM/ was downregulated (Figures 1c and 3b). Consistently, the kinase ATR-CHK1/CHK2 pathway are removed. CDC25 is activated and activity of CDK1 after DNA damage was slightly increased after WEE1 inhibited (in part through regulation by PLK1) to allow dephosphorylation and activation of CDK1. We found that MYT1 is MYT1 depletion (Figure 5c). Although effective depletion of MYT1 also inactivated during checkpoint recovery, possibly also depen- by the siRNAs was generally observed (for example, Figures 1c and dent on PLK1. Downregulation of MYT1 accelerates checkpoint 3b and Supplementary Figure S5C), depletion was likely to be recovery and mitotic entry. incomplete. Indeed, residual MYT1 kinase activity could be detected in vitro after siMYT1 transfection (Figure 7d). But importantly, such downregulation of MYT1 was able to reduce the long-term survival of irradiated cells, as indicated by clonogenic assays (Figure 6a) and growth in nude mice leading to a significant impact on long-term survival (Figure 6). In (Figures 6c and d). this connection, it would be interesting to examine the role of Conceptually, the checkpoint is inactivated once the damaged MYT1 in checkpoint recovery in the absence of checkpoint- DNA is repaired, allowing the cell cycle to resume. To study bypassing agents such as UCN-01 or MK1775. However, checkpoint recovery in the entire population, chemicals that these types of analysis are difficult due to the large variation of abrogated the checkpoint, including inhibitors of CHK1 (UCN-01) the time required for checkpoint recovery (or repair) between and WEE1 (MK1775) were generally used. While these models individual cells. have their limitations, they enable us to study checkpoint recovery It is known that PLK1 has an important role in checkpoint synchronously, without being affected by the heterogeneity of recovery.4 Indeed, we confirmed that depletion of PLK1 individual cells with different levels of DNA damage. Using these significantly delayed mitotic entry during checkpoint recovery. models, we found that depletion of MYT1 induced precocious Interestingly, mitotic entry time was shortened when PLK1 and entry into mitosis during checkpoint recovery. We demonstrated MYT1 were co-depleted, suggesting that MYT1 is downstream of this using time-lapse microscopy (Figure 2), flow cytometry PLK1 in the checkpoint recovery pathway (Figures 2 and 3). This is (Figure 3a), immunoblotting of mitotic markers (Figure 3b) and consistent with studies indicating that PLK1 can phosphorylate CDK1 kinase activity (Figure 5d). The conclusion was further the C-terminus of MYT1.9 In agreement with these, we found that substantiated using another siMYT1 (Supplementary Figure S5) as MYT1 was activated during checkpoint activation and inactivated well as a different cell line (Supplementary Figure S6). Moreover, during checkpoint recovery (Figures 7b and e). Furthermore, MYT1 depletion of MYT1 also accelerated mitotic entry when the isolated from checkpoint-activated cells could be inhibited by checkpoint was abrogated with MK1775 (Figure 4), indicating that PLK1 in vitro (Figure 7c). MYT1 had an independent role from WEE1. The model depicted in DNA damage followed by checkpoint abrogation has been Figure 8 illustrates the role of MYT1 in checkpoint recovery. developed as a strategy for anti-cancer therapies. Forcing The underlying mechanism of the accelerated checkpoint damaged cells into premature mitosis frequently triggers cell recovery in MYT1-depleted cells is likely to be due to the lowering death through mitotic catastrophe. However, some checkpoint- of threshold for CDK1 activation. MYT1 depletion reduced CDK1 abrogated cells can remain viable and progress into , inhibitory phosphorylation (Figures 1c and 3b) and increased which may contribute to further genome instability. Our previous CDK1 kinase activity (Figure 5c) after DNA damage. Although studies reveal that the extent of DNA damage and the sufficient to affect checkpoint recovery, the lowering of CDK1 effectiveness of the spindle-assembly checkpoint are pivotal inhibitory phosphorylation was clearly unable to promote mitosis determinants of mitotic catastrophe after checkpoint abroga- during checkpoint activation (at least for the first 36 h after DNA tion.20 In this study, we found that downregulation of MYT1 damage, Supplementary Figure S3). In contrast, targeting WEE1 shortened the time between checkpoint abrogation and mitotic triggered mitosis immediately (Figure 4), confirming WEE1 had a entry. Importantly, this also increased the level of subsequent cell more central role in keeping CDK1 inactive during checkpoint death (Figure 2b). An implication of these results is that MYT1 may activation. The precocious checkpoint recovery after the down- be a useful target for anti-cancer therapies. In agreement with this, regulation of MYT1 is likely to undermine effective DNA repair, combined treatments with IR and siMYT1 reduced tumor growth

Oncogene (2013) 4778 – 4788 & 2013 Macmillan Publishers Limited MYT1 in checkpoint recovery JPH Chow et al 4787 in xenograft models (Figure 6). MYT1 is a particularly attractive syringe with 15 strokes. After centrifugation at 4 1C at 3200 r.p.m. for 5 min, target because it is not required for normal cell cycle progression the pellet was collected as the nuclear fraction. The supernatant was (Figure 1). Cell cycle components that are essential for normal cell centrifuged again at 41C at 14 000 r.p.m. for 10 min. The supernatant and cycle progression (for example, WEE1 or PLK1) may be too toxic pellet were collected as the cytosol and membrane fractions, respectively. for normal cells and unsuitable for use in therapies. Antibodies and immunological methods In conclusion, downregulation of MYT1 reduces the inhibitory 33 34 26 phosphorylation of CDK1 during checkpoint activation. This leads Monoclonal antibodies against b-actin, cyclin A2 and cyclin B1 were to precocious mitotic entry during checkpoint recovery and obtained from sources as described previously. Polyclonal antibodies against gH2AX (Upstate Biotechnology Lake Placid, NY, USA), phospho- increase of cytotoxicity after checkpoint abrogation. These data CDK1Thr14 (Abcam,Cambridge,UK),phospho-CDK1Tyr15 ,phospho-CDK1Thr161 underscore the function of MYT1 in checkpoint recovery and ( Technology, Beverly, MA, USA), phospho-histone H3Ser10, implicate MYT1 as a potential target for anti-cancer therapies. PLK1 and MYT1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were obtained from the indicated suppliers. Monoclonal antibodies against CDC25C, and WEE1 were obtained from Santa Cruz Biotechnology. MATERIALS AND METHODS Immunoblotting and immunoprecipitation were performed as described.35 Materials Cells were prepared for immunofluorescence microscopy as previously 20 All reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA) unless described. stated otherwise. Kinase and phosphatase assays The expression (in bacteria) and purification of GST-CDC25A and GST-PLK1 Cell culture and cell growth analysis with glutathione-agarose chromatography were as described previously.35 The HeLa cell line used in this study was a clone that expressed the Tet For treatment with GST-CDC25A, immunoprecipitates were incubated with 25 26 20 . HeLa and U2OS expressing histone H2B-GFP and HeLa cells 1 mg of GST-CDC25A in 10 ml of buffer A (10 mM HEPES pH 7.2, 25 mM KCl, expressing FLAG–cyclin B127 were generated as previously described. Cells 10 mM NaCl, 1.1 mM MgCl2, 0.1 mM EDTA, 0.1 mM DTT) at 25 1C for 30 min. were propagated in DMEM supplemented with 10%(v/v) calf serum (for After washing in kinase buffer, the histone H1 kinase activity was assayed HeLa) or fetal bovine serum (for U2OS) (Invitrogen, Carlsbad, CA, USA) and as described previously.35 Phosphorylation was quantified with a 50 U/ml penicillin-streptomycin (Invitrogen) in a humidified incubator at PhosphorImager (Bio-Rad, Hercules, CA, USA). To assay MYT1 activity, 37 1Cin5%CO2. Unless stated otherwise, cells were treated with the HeLa cells stably expressing FLAG–cyclin B1 were treated with nocodazole following reagents at the final concentration: hydroxyurea (1.5 mM), MG132 for 16 h before harvested by mechanical shake-off. Lysates were prepared (10 mM), MK1775 (100 nM), nocodazole (0.1 mg/ml) and UCN-01 (100 nM). and 300 mg was subjected to immunoprecipitation using beads conjugated 22 28 Double thymidine synchronization, trypan blue analyses and with anti-FLAG antibodies. The FLAG–cyclin B1–CDK1 complexes were 29 clonogenic survival assays were performed as previously described. eluted by incubation with 0.2 mg of FLAG peptides in 10 ml of buffer A. The eluent was incubated with the membrane fraction in 50 ml of MYT1 kinase buffer (50 mM Tris–Cl pH 8.0, 10 mM MgCl2,50mM ATP, 1 mM DTT) DNA constructs 32 supplemented with 0.2 mlof(g- P)-ATP (for radioactive assay) and 10 mM MYT1 fragment was amplified by PCR with 50-GCCAGCCATGGTA of RO3306 at 30 1C for 1 h. For PLK1 treatment, the membrane fractions GAACGGCCTCCTGCACT-30 and 50-GTTAAAAGTGCAGAGGCAGAGTCGGA were incubated with 1 mg of GST or GST-PLK1 in 20 ml of MYT1 kinase TCCTCAGG-30 using -MYT1 in pcDNA3 (a gift from Helen Piwnica- 30 buffer at 25 1C for 30 min. For lPPase treatments, immunoprecipitates or Worms) as a template; the PCR fragment was partially digested with Nco I 31 membrane fractions were washed once with buffer (50 mM HEPES pH 7.5, and BamH I and ligated into pUHD-P3T to generate FLAG–MYT1 in 2mM MnCl , 100 mM EGTA, 5 mM DTT and 0.01% BRIJ-35) and incubated pUHD-P3T. To construct GST-PLK1, PLK1 was amplified by PCR with 50-G 2 with 400 U of lPPase in 10 ml of buffer at 301C for 1 h. CTGAATTCGCATGAGTGCTGCAG-30 and 50-GGGCTCGAGTTAGGAGGCCTTG AG-30 using myc-PLK1 in pRcCMV (a gift from Katsumi Yamashita) as a template; the PCR product was digested with EcoR I and Xho I and ligated Tumor xenografts into pGEX-KG. GST-CDC25A was described previously.32 The experimental protocol was evaluated and approved by the Animal Care Committee, HKUST. HeLa cells (2  106) were injected subcutaneously into both sides of the dorsa of 4–8 week-old female BALB/c nude mice. RNA interference Three animals per group were used in each experiment. Tumors were Stealth siRNAs targeting WEE1 (50-CCUCAGGACAGUGUCGUCGUAGAAA-30) regularly measured using a Vernier caliper. Volume was calculated and PLK1 (50-CAGCCUGCAGUACAUAGAGCGUGAU-30) were obtained from according to the formula: p/6  length  width2.36 Mice were killed when Invitrogen. Two siRNAs targeting MYT1 (50-CCUGGAUUCUCCCUCAAGAT tumors reached 1000 mm3. T-30 and 50-GUGACAUCAACUCAGAGCCTT-30) were obtained from RibioBio (Guangzhou, China). Unless stated otherwise, the first MYT1 siRNA was used. Cells were transfected with siRNA using Lipofectamine RNAiMAX CONFLICT OF INTEREST (Invitrogen) according to the manufacturer’s instructions. The authors declare no conflict of interest.

Ionizing radiation IR was delivered as described previously.29 ACKNOWLEDGEMENTS We thank Helen Piwnica-Worms and Katsumi Yamashita for generous gifts of Live-cell imaging reagents. This work was supported in part by grants AOE-MG/M-08/06 and HKU7/ CRG/09 to RYCP. Cells were seeded onto 24-well plates and imaged as described previously.29 Data acquisition was carried out at 5 min per frame. REFERENCES Flow cytometry 1 Fung TK, Poon RYC. A roller coaster ride with the mitotic cyclins. Semin Cell Dev Flow cytometry analysis after propidium iodide staining was performed as Biol 2005; 16: 335–342. described previously.28 2 Booher RN, Holman PS, Fattaey A. Human Myt1 is a cell cycle-regulated kinase that inhibits Cdc2 but not Cdk2 activity. J Biol Chem 1997; 272: 22300–22306. 3 Parker LL, Piwnica-Worms H. Inactivation of the p34cdc2-cyclin B complex by the Fractionation human WEE1 tyrosine kinase. Science 1992; 257: 1955–1957. HeLa cells growing on 100-mm plates were washed with 10 ml of PBS and 4 van Vugt MA, Bras A, Medema RH. Polo-like kinase-1 controls recovery from a G2 harvested by a cell scraper in 1 ml of PBS. The cells were collected by DNA damage-induced arrest in mammalian cells. Mol Cell 2004; 15: 799–811. centrifugation at 5000 r.p.m. for 150 s. The pellet were resuspended in 5 Watanabe N, Arai H, Iwasaki J, Shiina M, Ogata K, Hunter T et al. Cyclin-dependent 500 ml of lysis buffer (10 mM PIPES pH 7.4, 10 mM KCl, 2 mM MgCl2,1mM kinase (CDK) phosphorylation destabilizes somatic Wee1 via multiple pathways. DTT) and incubated on ice for 15 min. The cells were passed through a 25G Proc Natl Acad Sci USA 2005; 102: 11663–11668.

& 2013 Macmillan Publishers Limited Oncogene (2013) 4778 – 4788 MYT1 in checkpoint recovery JPH Chow et al 4788 6 Watanabe N, Arai H, Nishihara Y, Taniguchi M, Watanabe N, Hunter T 22 Chow JP, Poon RYC, Ma HT. Inhibitory phosphorylation of cyclin-dependent et al. M-phase kinases induce phospho-dependent ubiquitination of somatic kinase 1 as a compensatory mechanism for mitosis exit. Mol Cell Biol 2011; 31: Wee1 by SCFbeta-TrCP. Proc Natl Acad Sci USA 2004; 101: 4419–4424. 1478–1491. 7 Katayama K, Fujita N, Tsuruo T. Akt/ B-dependent phosphorylation 23 Krek W, Nigg EA. Mutations of p34cdc2 phosphorylation sites induce premature and inactivation of WEE1Hu promote cell cycle progression at G2/M transition. mitotic events in HeLa cells: evidence for a double block to p34cdc2 kinase Mol Cell Biol 2005; 25: 5725–5737. activation in vertebrates. EMBO J 1991; 10: 3331–3341. 8 Wells NJ, Watanabe N, Tokusumi T, Jiang W, Verdecia MA, Hunter T. The 24 Wang Y, Decker SJ, Sebolt-Leopold J. Knockdown of Chk1, Wee1 and Myt1 by C-terminal domain of the Cdc2 inhibitory kinase Myt1 interacts with Cdc2 com- RNA interference abrogates G2 checkpoint and induces . Cancer Biol plexes and is required for inhibition of G(2)/M progression. J Cell Sci 1999; 112: Ther 2004; 3: 305–313. 3361–3371. 25 Yam CH, Siu WY, Lau A, Poon RYC. Degradation of does not require its 9 Nakajima H, Toyoshima-Morimoto F, Taniguchi E, Nishida E. Identification of a phosphorylation by CDC2 and cyclin-dependent kinase 2. J Biol Chem 2000; 275: consensus motif for Plk (Polo-like kinase) phosphorylation reveals Myt1 as a Plk1 3158–3167. substrate. J Biol Chem 2003; 278: 25277–25280. 26 Chan YW, Ma HT, Wong W, Ho CC, On KF, Poon RYC. CDK1 inhibitors antagonize 10 Stumpff J, Duncan T, Homola E, Campbell SD, Su TT. Wee1 the immediate apoptosis triggered by spindle disruption but promote apoptosis kinase regulates Cdk1 and mitotic entry during embryogenesis. Curr Biol 2004; 14: following the subsequent rereplication and abnormal mitosis. Cell Cycle 2008; 7: 2143–2148. 1449–1461. 11 Nakajima H, Yonemura S, Murata M, Nakamura N, Piwnica-Worms H, Nishida E. 27 Chan YW, Chen Y, Poon RYC. Generation of an indestructible cyclin B1 by caspase- Myt1 protein kinase is essential for Golgi and ER assembly during mitotic exit. 6-dependent during mitotic catastrophe. Oncogene 2009; 28: 170–183. J Cell Biol 2008; 181: 89–103. 28 Siu WY, Arooz T, Poon RYC. Differential responses of proliferating versus quiescent 12 Nakajo N, Yoshitome S, Iwashita J, Iida M, Uto K, Ueno S et al. Absence cells to adriamycin. Exp Cell Res 1999; 250: 131–141. of Wee1 ensures the meiotic cell cycle in Xenopus oocytes. Dev 2000; 14: 29 Chen Y, Chow JP, Poon RY. Inhibition of Eg5 acts synergistically with checkpoint 328–338. abrogation in promoting mitotic catastrophe. Mol Cancer Res 2012; 10: 626–635. 13 Ruiz EJ, Hunt T, Nebreda AR. Meiotic inactivation of Xenopus Myt1 by CDK/ 30 Liu F, Stanton JJ, Wu Z, Piwnica-Worms H. The human Myt1 kinase preferentially XRINGO, but not CDK/cyclin, via site-specific phosphorylation. Mol Cell 2008; 32: phosphorylates Cdc2 on threonine 14 and localizes to the endoplasmic reticulum 210–220. and Golgi complex. Mol Cell Biol 1997; 17: 571–583. 14 Okumura E, Fukuhara T, Yoshida H, Hanada Si S, Kozutsumi R, Mori M et al. Akt 31 Ma HT, Tsang YH, Marxer M, Poon RYC. Cyclin A2-cyclin-dependent kinase 2 inhibits Myt1 in the signalling pathway that leads to meiotic G2/M-phase tran- cooperates with the PLK1-SCFbeta-TrCP1-EMI1-anaphase-promoting complex/ sition. Nat Cell Biol 2002; 4: 111–116. cyclosome axis to promote genome reduplication in the absence of mitosis. Mol 15 Chen Y, Poon RYC. The multiple checkpoint functions of CHK1 and CHK2 in Cell Biol 2009; 29: 6500–6514. maintenance of genome stability. Front Biosci 2008; 13: 5016–5029. 32 Chow JP, Siu WY, Ho HT, Ma KH, Ho CC, Poon RYC. Differential contribution of 16 Jin P, Gu Y, Morgan DO. Role of inhibitory CDC2 phosphorylation in radiation- inhibitory phosphorylation of CDC2 and CDK2 for unperturbed cell cycle control induced G2 arrest in human cells. J Cell Biol 1996; 134: 963–970. and DNA integrity checkpoints. J Biol Chem 2003; 278: 40815–40828. 17 O’Connell MJ, Raleigh JM, Verkade HM, Nurse P. Chk1 is a wee1 kinase in the G2 33 Chan YW, On KF, Chan WM, Wong W, Siu HO, Hau PM et al. The kinetics of p53 DNA damage checkpoint inhibiting cdc2 by Y15 phosphorylation. EMBO J 1997; activation versus cyclin E accumulation underlies the relationship between the 16: 545–554. spindle-assembly checkpoint and the postmitotic checkpoint. J Biol Chem 2008; 18 van Vugt MA, Medema RH. Getting in and out of mitosis with Polo-like kinase-1. 283: 15716–15723. Oncogene 2005; 24: 2844–2859. 34 Yam CH, Siu WY, Kaganovich D, Ruderman JV, Poon RYC. Cleavage of cyclin A at 19 Karlsson-Rosenthal C, Millar JB. Cdc25: mechanisms of checkpoint inhibition and R70/R71 by the bacterial protease OmpT. Proc Natl Acad Sci USA 2001; 98: recovery. Trends Cell Biol 2006; 16: 285–292. 497–501. 20 On KF, Chen Y, Ma HT, Chow JP, Poon RYC. Determinants of mitotic catastrophe 35 Poon RYC, Toyoshima H, Hunter T. Redistribution of the CDK inhibitor p27 on abrogation of the G2 DNA damage checkpoint by UCN-01. Mol Cancer Ther between different cyclin.CDK complexes in the mouse fibroblast cell cycle 2011; 10: 784–794. and in cells arrested with lovastatin or ultraviolet irradiation. Mol Biol Cell 1995; 6: 21 Mizuarai S, Yamanaka K, Itadani H, Arai T, Nishibata T, Hirai H et al. Discovery of 1197–1213. expression-based pharmacodynamic biomarker for a p53 context-specific 36 Tomayko MM, Reynolds CP. Determination of subcutaneous tumor size in athymic anti-tumor drug Wee1 inhibitor. Mol Cancer 2009; 8:34. (nude) mice. Cancer Chemother Pharmacol 1989; 24: 148–154.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene (2013) 4778 – 4788 & 2013 Macmillan Publishers Limited